Dwarf galaxies in the ΛCDM model

Mario G. AbadiObservatorio Astronómico

Universidad Nacional de Córdoba (UNC) Instituto de Astronomía Teórica y Experimental CONICET-UNC

February 4-7 , 2013Cosmology, Large Scale Structure and First Object

USP Conference, Sao Paulo, Brazil

Collaborators:

Alejandro Benítez-Llambay (Córdoba)Ismael Ferrero (Córdoba) Julio Navarro (Victoria)Laura Sales (Garching)Sebastian Gurovich (Córdoba)Stefan Gottloeber (Potsdam) Gustavo Yepes (Madrid)Yehuda Hoffman (Jerusalem)Matthias Steinmetz (Potsdam)

Outline

Tuesday Feb. 05Part II – Structure of the Universe: Theory, Simulations and ObservationsII.4 Comparison of simulations with observations:

“Dwarf galaxies in the ΛCDM model”

There seems to be some tension between observations and simulations in dwarf galaxy scales.

I) Observations II) Simulations

Galaxy Stellar Mass Function

~50000 galaxies from Sloan Digital Sky Survey with redshift

0.0033 <z<0.05 by Baldry et al 2008

Log Stellar Mass [Msolar]

Num

ber D

ensi

ty o

f Gal

axie

s pe

r Log

mas

s in

terv

al

[dex

^-1

Mpc

^-3]

Galaxy Stellar Mass Function

~50000 galaxies from Sloan Digital Sky Survey with redshift

0.0033 <z<0.05 by Baldry et al 2008

Log Stellar Mass [Msolar]

Num

ber D

ensi

ty o

f Gal

axie

s pe

r bin

[dex

^-1

Mpc

^-3]

MW Stellar mass5x10^10 Msolar

Dark Matter Halo Mass Function

Millennium series of Dark Matter only simulations (Springel et al 2005 & Angulo et al 2012)

Dark Matter Halo Mass

Dark Matter Halo Mass Function

Angulo et al 2012

Dark Matter Halo Mass

Millennium series of Dark Matter only simulations (Springel et al 2005 & Angulo et al 2012)

Galaxy and Halo Mass Function

At the faint end, the dark matter halo mass function is much steeper than the galaxy stellar mass function

What is the relation between galaxy stellar mass and dark matter halo mass in order to reconcile these two distributions?

GalaxiesBaldry et al 2008

Log Mass [Msolar]

Milennium Dark Matter Halos (rescaled)

Galaxy Stellar Mass vs Halo Mass

Galaxy–Halo mass relation assuming that dark haloes ranked by their mass are matched to galaxies ranked by stellar mass. The first most massive halo corresponds to the first most massive galaxy, an so on.

Log Dark Matter Halo Mass

Log

Stel

lar M

ass

[Mso

lar]

Abundance MatchingGuo et al 2010

Universal Baryon Fraction fbar~0.171

Galaxy Stellar Mass vs Halo Mass

Very steep relation at the low halo mass end

Essentially no galaxies with Mgal>10^6 Msolar should form in halos with mass below a threshold of 10^10 Msolar

Log Dark Matter Halo Mass

Log

Stel

lar M

ass

[Mso

lar]

Abundance MatchingGuo et al 2010

Universal Baryon Fraction fbar~0.171

Galaxy Stellar Mass vs Halo Mass

More sophisticated semi-analytic models of galaxy formation (Guo et al. 2011, green triangles) also show very similar results

Are this masses consistent with masses estimated from kinematic data?

Log Dark Matter Halo Mass

Log

Stel

lar M

ass

[Mso

lar]

Universal Baryon Fraction fbar~0.171

Abundance MatchingGuo et al 2010

Galaxy Stellar Mass vs Halo Mass

Abundance MatchingGuo et al 2011

Log Dark Matter Halo Mass

Log

Stel

lar M

ass

[Mso

lar]

Milky Way galaxy Stellar mass ~5x10^MsolarDark matter halo mass ~1.5x10^12 Msolar

Milky Way dSphs (Boylan-Kolchin 2011a,b)

Isolated dwarfs HI, tidal interactions negligible and more extended rotation curves

Milky Way

Dwarfs: UGC7559 & SDIG

Ferrero et al 2012

Abundance matching prediction

Observed Rotation Curves

Dwarf Spheroidals of the Milky Way

Dwarf spheroidal satellites of the Milky Way seems to live in halos with 12 km/s < Vmax< 24 km/s. (~10^10 Msolar)

(~10^9 Msolar)

Observational Sample

Outermost Measured Rotational Velocity

Ste

llar M

ass

Stellar Mass versus Rotational Velocity (“Tully-Fisher” relation) for a sample of nearby galaxies compiled from the literature.

Stellar masses are taken from each paper (or estimated from absolute magnitudes and colors, only a few cases).

Ferrero et al 2012

Measured versus Predicted Velocity

Abundance Matching Predicted Velocity

Mea

sure

d V

eloc

ity

Massive galaxies seem to be in good agreement with the abundance-matching model.

Galaxies with stellar masses below ~3x10^7 Msolar (magenta, red and green) symbols have velocities that fall systematically below the expected ~30 km/s corresponding to a halo mass of ~10^10 Msolar

Ferrero et al 2012

(Outermost) Velocity vs Radius

The outermost point rotational velocity for a sample of dwarf galaxies compiled from the literature.

Abundance-matching suggest that all points should lie on or above the shaded area labeled M200 = 10^10 Msolar. This is clearly not the case.

If there is a minimum halo mass for dwarf galaxy formation, it cannot be much higher than 5x10^8 M.∼

Ferrero et al 2012

Galaxy Stellar Mass vs Halo Mass

Average galaxy mass as a function of halo mass for our compiled sample.

It is clearly shallower than either the abundance-matching model resultsof Guo et al. (2010)or the semi-analytic model results of Guo et al. (2011).

Ferrero et al 2012

Galaxy Stellar Mass vs Halo Mass

These results agree with a similar analysis done by Oh et al 2011.

Caveats

1) Incorrect interpretation of the data: rotational velocity of neutral gas in dwarf irregulars is not a direct measure of its circular velocity. Gas pressure, non-circular motions and gas velocity dispersion corrections should be taken into account.

2) Baryonic effects: supernova driven gas blowouts (e.g. Navarro et al. 1996) or gravitational fluctuations created by star-forming regions (Pontzen & Governato 2012) might reduce the dark matter content of dwarf galaxies and alleviate the problem.

3) A mechanism to select a small fraction of low-mass haloes to be galaxy hosts while leaving dark the vast majority of systems of comparable (or even higher) mass.

CLUES Simulations

http://www.clues-project.org

CLUES Simulation

A cosmological numerical simulation that matches the Local Group nearby large-scale structure and the relative positions, stellar masses and morphology of its 3 main spirals: Milky Way (MW), Andromeda (M31) and Triangulum (M33).

SPH-GADGET2 (Springel 2005) code with cosmic ionizing UV background, star formation, Supernova feedback and isotropic winds.

Zoom-in technique with a Low res region = box of 64/h Mpc on a sideHigh res region = roughly spherical of 2/h Mpc radius

Initially, 53 millon gas plus dark matter particles mgas~6x10^4 Msolar and mdark~3.5x10^5 MsolarSoftening 0.14 kpc

Gas2Mpc/h 50 kpc/h

Image credit K. Riebe

Dark Matter

1.3 Mpc/h per sideG. Yepes

Star Formation Rate S

tar F

orm

ation

Rat

e

Galaxy Stellar Mass

We exclude form our sample all satellite galaxies and we focus only on central galaxies of halos with M>10^10 Msolar.Galaxy properties are computed within the galactic radius rgal=0.15 r200

Simulated Local Group galaxiesshow a wide range of star formation rates. Two groups are distinguishable: 1) Those that are forming stars at rates roughly comparable to their past average (green triangles)2) Those where star formation haslargely ceased (red circles)

Benitez-Llambay et al 2013

Age Distribution

Galaxies with on-going star formation Galaxies that stopped forming stars

Benitez-Llambay et al 2013

Baryonic Content Ba

ryon

ic (G

as +

Sta

rs) M

ass

insi

de V

irial

Rad

ius

Halo Virial Mass

Simulated galaxies where star formation has largely ceased (red circles) have lost most of their baryons.None of these galaxies has retainedmore than ~20% of the baryons withintheir virial radius.

Benitez-Llambay et al 2013

Mass Evolution M

ass

insi

de V

irial

Rad

ius

Time in Gyrs

Evolution of mass within the virialRadius for galaxy number 30. Most of the baryons are lost overa short period of time 2<z<3.The lost of baryons is the main reasonwhy this galaxy has stopped forming stars.

Benitez-Llambay et al 2013

Gas Temporal Evolution

Z=5.08 Z=3.01 Z=2.51

Z=1.80 Z=0.75 Z=0.00

Benitez-Llambay et al 2013

Trajectories of galaxiesBenitez-Llambay et al 2013

Ram Pressure Gunn & Gott 1972

Image Credit: H. Crowl (Yale University) and WIYN/NOAO/AURA/NSFAbadi et al 1999

Ram Pressure

P ~ ρ V^2

Benitez-Llambay et al 2013

More ExamplesBenitez-Llambay et al 2013

Conclusions

Abundance matching implies that galaxy formation must become extremely inefficient below 10^10 Msolar in order to reconcile the galaxy stellar mass function with dark halo mass function on galactic scales.

Many of the galaxies in our sample have enclosed masses much lower than expected from haloes as massive as 10^10 Msolar

”Cosmic web stripping” enables the removal of baryons from low-mass halos without appealing to feedback or reionization.

May help to explain the scarcity of dwarf galaxies compared with the numerous low-mass halos expected in the CDM and the large diversity of star formation histories and morphologies characteristic of faint galaxies.

Further simulations and observational evidence needed.